remote sensing

Article
Assessment of Spatio-Temporal Patterns of Black
Spruce Bud Phenology across Quebec Based on
MODIS-NDVI Time Series and Field Observations
Siddhartha Khare 1, * , Guillaume Drolet 2 , Jean-Daniel Sylvain 2 , Maxime Charles Paré 1
and Sergio Rossi 1,3
1 Département des Sciences Fondamentales, Université du Québec à Chicoutimi, Québec,
QC G7H2B1, Canada; maxime_pare@uqac.ca (M.C.P.); sergio_rossi@uqac.ca (S.R.)
2 Direction de la recherche forestière Ministère des Forêts, de la Faune et des Parcs, Québec,
QC G1P3W8, Canada; Guillaume.Drolet@mffp.gouv.qc.ca (G.D.);
Jean-Daniel.Sylvain@mffp.gouv.qc.ca (J.-D.S.)
3 Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, Guangdong Provincial
Key Laboratory of Applied Botany, South China Botanical Garden, Chinese Academy of Sciences,
Guangzhou 510650, China
* Correspondence: siddhartha.khare1@uqac.ca




Received: 16 October 2019; Accepted: 20 November 2019; Published: 22 November 2019


Abstract: Satellite remote sensing is a widely accessible tool to investigate the spatiotemporal
variations in the bud phenology of evergreen species, which show limited seasonal changes in canopy
greenness. However, there is a need for precise and compatible data to compare remote sensing time
series with field observations. In this study, fortnightly MODIS-NDVI was fitted using double-logistic
functions and calibrated using ordinal logit models with the sequential phases of bud phenology
collected during 2015, 2017 and 2018 in a black spruce stand. Bud break and bud set were spatialized
for the period 2009–2018 across 5000 stands in Quebec, Canada. The first phase of bud break and
the last phase of bud set were observed in the field in mid-May and at the beginning of September,
when NDVI was 80.5% and 92.2% of its maximum amplitude, respectively. The NDVI rate of change
was estimated at 0.07 in spring and 0.04 in autumn. When spatialized on the black spruce stands,
bud break was detected earlier in the southwestern regions (April–May), and later in the northeastern
regions (mid to end of June). No clear trend was observed for bud set, with different patterns
being detected among the years. Overall, the process bud break and bud set lasted 51 and 87 days,
respectively. Our results demonstrate the potential of satellite remote sensing for providing reliable
timings of bud phenological events using calibrated NDVI time series on wide regions that are remote
or with limited access.

Keywords: remote sensing; bud break; bud burst; bud set; boreal forest; NDVI

1. Introduction
Plant phenology is the study of recurring life cycle events, such as growth reactivation and
dormancy, leaf emergence and senescence, and flowering. Phenology is considered as a sensitive
bio-indicator of climate and its changes [1–3]. Over the past three decades, phenology studies have
been conducted using field-based observations, bioclimatic models, near-surface remote sensing
and satellite remote sensing based techniques [4–7]. Researchers have applied different approaches
depending on the question addressed and the spatial scale of investigation. Satellite data provide
wide coverage with varying temporal, spectral and spatial resolutions. Remote sensing is therefore
a flexible, reliable and widely recognized tool to study phenology in a number of ecosystems, from

Remote Sens. 2019, 11, 2745; doi:10.3390/rs11232745 www.mdpi.com/journal/remotesensing

Remote Sens. 2019, 11, 2745 2 of 16

forests to grasslands, which may be too difficult using other data collection methods [8–10]. Current
sensors on board satellite platforms record spectral signals that can be used to monitor seasonal and
interannual variations in vegetation cover and determine the timings of phenological events and the
growing season across the landscape [11,12].
Imagery from the Moderate Resolution Imaging Spectroradiometer (MODIS) on board the
Terra and Aqua satellites is widely used to study vegetation phenology [4,13]. Among the available
tools, the Normalized Difference Vegetation Index (NDVI) [14,15] is a reliable spectral index for the
reconstruction of phenological transitions of different vegetation types, including croplands, forests
and grasslands, because it is related to the amount of green-leaf biomass [13,16,17]. NDVI is considered
as a potential remote sensing proxy to investigate the effect of climate on vegetation phenology at
regional to continental scales due to its close relationship with plant activity [18,19].
In the last two decades, substantial advances have been made in predicting length and start/end
dates of the growing season from both field observations and satellite remote sensing data [20–22].
However, the scarcity of ground observations in some locations is a challenge for studies of growth
patterns over large or remote regions of the world [23,24]. Another constraint exists in stands of
evergreen species, which show only small seasonal variations in canopy greenness compared to
deciduous species [21,25,26]. Phenology and growing season in North America have been studied
extensively using the NDVI at medium (500 m) and coarse (1 and 8 km) spatial resolutions [22,27,28].
However, direct observations in the field confirming the results from remote sensing in remote areas
remain scarce.
To address these issues, we propose an innovative statistical approach to calibrate MODIS NDVI
time series and demonstrate how it can be used to describe annual changes in black spruce [Picea mariana
(Mill.) bud phenology across a large portion of the coniferous forest in a study area in Quebec, Canada.
This approach provides a robust alternative to field-based monitoring, especially in areas where field
observations are difficult due to the remoteness or extreme weather conditions.

2. Materials and Methods

2.1. Study Area and Field Observations Site

The overall study area is located in Quebec, Canada, and covers nearly 800,000 km2 between
45.5◦ –55◦ N and 64◦ –80◦ W. Using the Quebec 1:20k forest map from the Quebec Government [29],
we selected 5000 polygons representing stands dominated by black spruce trees (>75%) and that had
not been disturbed over the past 30 years (Figure 1a).
Field measurements in this study were taken in a black spruce stand of the Forêt d’Enseignement
et de Recherche Simoncouche (48◦ 220 N, 71◦ 250 W, 338 m a.s.l.) in Quebec, Canada. The study site is
located within the balsam fir (Abies balsamea L. Mill.)-white birch (Betula papyrifera Marsh.) bioclimatic
domain, at the southern border of the boreal forest (Figure 1b). The site has a typical boreal climate
characterized by short cool summers and cold harsh winters. Mean the annual temperature is 0.9 ◦ C,
with absolute minimum and maximum temperatures of −36.7 ◦ C in January–February, and 31.1 ◦ C in
July, respectively. Mean annual precipitation is 1162 mm [30].

2.2. Identification of the Phenological Phases

We collected observations of bud phenology on 105 black spruce saplings each week from May
to October during 2015, 2017 and 2018. We collected data on saplings due to the need to perform
reliable observations directly on the buds, which would have been difficult to perform on tall mature
trees. All phenological phases of bud break and bud set were recorded according to [31]. Six phases
were defined for bud break: open bud (B1), with a pale spot at the tip of the bud; elongated bud
(B2), with lengthening brown scales; swollen bud (B3), with smooth and pale-coloured scales but no
visible needles; translucent bud (B4), with needles visible through the scales; split bud (B5), with open
scales but needles still clustered; and exposed shoot (B6), with needles completely emerged from the
 Remote Sens. 2019, 11, 2745 3 of 16

Remote Sens. 2019, 11, x FOR PEER REVIEW 3 of 17

surrounding scales and spreading outwards. Five phases of bud set were defined: white bud (S1),
with the presence of a white bud; beige bud (S2), with beige scales around the bud; brownish bud (S3),
(S3),
withwith a significant
a significant increaseininvolume;
increase volume;brown
brown bud (S4),
(S4), with
withneedles
needlesstarting
startingtoto
spread outwards;
spread outwards; and
and spread
spread needles(S5),
needles (S5),with
withthe
theneedles
needles in
in the
the whorl spreading
spreadingoutwards.
outwards.

Figure
Figure1. Location of the
1. Location of 5000 blackblack
the 5000 spruce stands
spruce in Quebec,
stands Canada
in Quebec, (a) selected
Canada from two
(a) selected ISIN
from tiles
two ISIN tiles
(h13 v03 and h13 v04, not shown) of MODIS, and the stand where phenological phases were
(h13 v03 and h13 v04, not shown) of MODIS, and the stand where phenological phases were measured measured
(b).(b).
TheThe
zoomed-in
zoomed-inareaarea
shows the 40
shows black
the spruce
40 black polygons
spruce (in red(in
polygons color) used for
red color) calibration
used within within
for calibration
thethe
buffer of 25 km 2 around the field observation site.
2
buffer of 25 km around the field observation site.

2.3. MODIS
2.3. Data
MODIS Data

WeWeacquired
acquiredtime
timeseries
series ofof 250
250 m mTerra
TerraModerate
ModerateResolution
Resolution Imaging
ImagingSpectroradiometer
Spectroradiometer (MODIS)
Vegetation
(MODIS) Indices (MOD13Q1,
Vegetation version 6) from
Indices (MOD13Q1, the NASA
version Earthdata
6) from the website
NASA (https://earthdata.nasa.gov).
Earthdata website
MOD13Q1 is derived from atmospherically corrected bi-directional
(https://earthdata.nasa.gov). MOD13Q1 is derived from atmospherically corrected surface reflectance imagery and
bi-directional
surface reflectance imagery and contains vegetation index data as well as the pixel reliability layer [32].
contains vegetation index data as well as the pixel reliability layer needed for quality checking
We used
needed 16 day checking
for quality NDVI maximum
[32]. We usedcomposite
16 dayvalues in ISIN tiles
NDVI maximum h13v03 values
composite and h13v04in ISIN from
tiles2009 to
2018,and
h13v03 thus covering
h13v04 froma2009large portion
to 2018, thusofcovering
the eastern Quebec
a large portionboreal and temperate
of the eastern Quebec borealforests.
andIn total,
temperate forests. In
this represented total,
460 NDVI thisgranules
represented(230460
forNDVI granules
each ISIN (230
tile for forwhole
the each ISIN tile for
period). All the
thewhole
images were
period).
filteredAll
tothe images
exclude were filtered
unreliable NDVI to pixels
exclude unreliable
using a pixelNDVI pixels
reliability using
layer a pixel reliability
supplied by MODISlayer vegetation
supplied by MODISteam.
index production vegetation
MODISindexscenesproduction team. MODIS
were re-projected scenes wereLambert
to NAD83/Quebec re-projected to
projection and
NAD83/Quebec Lambert projection and stacked by band (i.e., vegetation index)
stacked by band (i.e., vegetation index) and by date of the composite period for each tile. and by date of the
composite period for each tile.
2.4. Assessing and Calibrating NDVI
2.4. Assessing and Calibrating NDVI
We selected 40 polygons with an area:perimeter ratio greater than 13 for pure black spruce stands
We selected
adjacent 40 to
or close polygons with an area:perimeter
the Simoncouche ratioThe
study site [29]. greater than
forest 13 for pure
polygons black
selected forspruce
all analyses
stands adjacent or close to the Simoncouche study site [29]. The forest polygons
included stands composed by >75% of black spruce. We used these polygons to derive selected forstand-level
all
analyses included stands composed by >75% of black spruce. We used these polygons to derive stand-
NDVI time series by calculating, for each NDVI image, the mean of all pixels located within the limits
level NDVI time series by calculating, for each NDVI image, the mean of all pixels located within the
Remote Sens. 2019, 11, 2745 4 of 16

of the polygons. NDVI phenology curves were then fitted for each year using a double-logistic model,
where time (t) was represented by Day Of the Year (DOY):
!
1 1
NDVI = min + (max − min) × + −1 (1)
1 + exp(−mS × (t − S)) 1 + exp(mA × (t − A))

Three pairs of coefficients were included: minimum and maximum NDVI (min and max), timings
of the inflection points (S and A), and rates of change at the inflection points (mS and mA) [20,33].
The estimated double-logistic functions were standardized (stdNDVI ) to the range 0–1 and compared to
field observations of phenological phases.
!
NDVI − min(NDVI )
stdNDVI = (2)
max(NDVI ) − min(NDVI )

We applied logit models for bud break (six phases, i.e., B1–B6) and bud set (five phases,
i.e., S1–S5) to estimate the probability of observing each sequential phenological phase along the
stdNDVI measurements. This procedure fits the proportional odds model to ordinal response data,
here represented by the sequential phases of bud phenology. The ordinal logit models were applied
using the LOGISTIC procedure in SAS [34]. We assessed the performance of the model three times,
each time using two years of bud phenology observations for model training and the remaining year for
validation. Observations and predictions were compared by using linear regression and the goodness
of fit (R2 ) and the root mean square error (RMSE) in prediction as performance metrics [35].

2.5. Spatializing Black Spruce Phenology

We again extracted NDVI values but this time for the period from 2009 to 2018 and for 5000
pure black spruce sites [29] distributed over two MODIS tiles covering our study area. Polygons
with a ratio area:perimeter > 13 were selected and for each polygon, NDVI pixels were extracted and
used to calculate mean polygon NDVI for each MODIS image. The area:perimeter threshold value
was used to ensure that we used mostly clumps of pixels and excluded long and narrow polygons,
this decreasing the risk of including mixed pixels in our dataset. We applied the thresholds estimated
by the above-mentioned models to estimate the beginning (B1) and ending (S5) of bud phenology.
We performed a hotspot and coldspot analysis using Getis-Ord Gi* [36] in ArcGIS 10.2 (Environmental
Systems Research Institute, Inc., Redlands, CA) to identify spatial clustering in phenological events [37].
We used the inverse distance weighted interpolation method to interpolate the resulting values over
the entire area [38].

3. Results

3.1. MODIS NDVI Data

Annually, NDVI showed a bell-shaped pattern with a slow increase in spring, followed by a rapid
increase and culmination in July and a decrease in autumn until reaching a minimum value in winter
(Figure 2). On average, NDVI ranged from 0.32 in winter and spring to 0.9 in summer. The double-logistic
function represented NDVI well during the whole season, including the asymmetry in the transitions
between spring, summer and autumn. The satisfactory fitting was confirmed by the analysis of the
residuals, which were uniformly distributed around zero for the three studied years. Only 3% of the
residuals exceeded the range between −0.2 and 0.2, confirming the model reliability (Figure 2).
Compared to 2017 and 2018, the year 2015 showed the highest NDVI (0.917) in summer (Table 1).
In 2015 and 2018, the inflection points (coefficients S and A) occurred at approximately the same time,
at the beginning of May (DOY 121 and 122) and the end of October (DOY 284 and 290), respectively.
In 2017, NDVI started to increase earlier (DOY 118) and decreased later (DOY 311) compared to the
other two years, which resulted in a 20% increase of the growing season (DOY 192) compared to 2017
Remote Sens. 2019, 11, 2745 5 of 16
Remote Sens. 2019, 11, x FOR PEER REVIEW 5 of 17

(DOY 162) and 2018 (DOY 167). On average, coefficients mS and mA were estimated at 0.07 and 0.04,
at 0.07 and demonstrating
respectively, 0.04, respectively, demonstrating
that thatofautumnal
autumnal rates rates
reduction of reduction
in NDVI in NDVI
were slower were
than slower in
increases
than increases in
the spring (Table 1). the spring (Table 1).

Figure
Figure 2. 2. NDVItime
NDVI timeseries
series(grey
(greycircles)
circles) and
and fitted double-logistic
double-logisticcurves
curves(solid
(solidblack line)
black forfor
line) three
three
years
years in in a black
a black spruceboreal
spruce borealstand
standof
ofQuebec,
Quebec, Canada.
Canada. Each
Eachday
dayofofthe
theyear
yearshows
shows the mean
the meanNDVI
NDVI
values
values forfor
4040 black
black sprucestands
spruce standsused
usedfor
formodel
model calibration.
calibration.

Table 1. Coefficients of the double-logistic function fitted to three years of MODIS NDVI values
Table 1. Coefficients
extracted for 40 pure of thespruce
black double-logistic function
boreal stands fitted
in Quebec, to three
Canada. minyears of MODIS
and max representNDVI values
minimum
extracted for 40 pure black spruce boreal stands in Quebec, Canada. min and max represent minimum
and maximum NDVI, S and A represent timings of the inflection points and rates of change at the
and maximum
inflection NDVI,
points S and Abyrepresent
represented mS and mA.timings of the inflection points and rates of change at the
inflection points represented by mS and mA.
Year
Coefficients Year
Coefficients 2015 2017 2018
min 20150.322 0.3252017 0.291 2018
min max 0.3220.917 0.8580.325 0.875 0.291
max S 0.917
121.04 0.858122.43
118.34 0.875
S A 121.04
283.78 118.34289.73
310.86 122.43
A 283.78 310.86 289.73
mS 0.06 0.08 0.07
mS 0.06 0.08 0.07
mA -0.04 -0.05 -0.05
mA −0.04 −0.05 −0.05

3.2. Comparing NDVI with Bud Phenology

3.2. Comparing NDVI with Bud Phenology
Bud break (B1–B6) and part of bud set (S1–S2) occurred under increasing (Figure 3).
Bud break
varied (B1–B6)
from 0.81and partduring
to 0.97 of bud bud setbreak
(S1–S2)
(i.e.,occurred at phasestd
under increasing
B1–B6), culminating S2.NDVI (Figure
Phases S3 to 3).
stdS5
NDVI varied from 0.81
occurred under decreasingto 0.97 during bud
, at values ranging between 0.92 and 1. Larger standard S3
break (i.e., B1–B6), culminating at phase S2. Phases
to deviations
S5 occurred wereunder
observed for budstd
decreasing break , at values
NDVIphases than forranging between
bud set, 0.92
except for andS5.
phase 1. The
Larger standard
probability
deviations were observed for bud break phases
of occurrence of each bud phenological phase according to than for bud set, except for phase S5. The probability
was estimated with ordinal logit
of models.
occurrenceModelof each bud phenological
validation produced satisfying estimationstoofstd
phase according the was estimated
timings
NDVI with ordinal
of the phenological phases logit
models.
(FigureModel
4). Thevalidation produced
results of the satisfying
model were estimations
significant, of thebytimings
as indicated the highofRthe phenological
2 (0.87) and a RMSE phases
of
12.85.4).
(Figure The regression
The results ofestimated
the model a slope of 0.97, which
were significant, was not statistically
as indicated R2 (0.87)from
by the highdifferent and one
a RMSE (p > of
The regression estimated a slope of 0.97, which was not statistically different from one (p > 0.05).
0.05).
12.85.
With stdNDVI < 0.70,
With < 0.70,
thethe probability
probability of observing
of observing a dormant
a dormant bud (phase
bud (phase B0) wasB0)0.94,
waswhich
0.94, which
reduced
at reduced
increasing at increasing
stdNDVI (Figure 5).(Figure 5). The probability
The probability of observing
of observing B1 exceeded
B1 exceeded dormant dormant
bud atbud at
stdNDVI
of 0.75, meaning that at this value, the buds were more
of 0.75, meaning that at this stdNDVI value, the buds were more likely to be active than dormant.likely to be active than
B1dormant.
culminated B1 culminated
at stdNDVI at of 0.79, reaching
of 0.79, reaching the highestthe highest probability
probability of occurring.
of occurring. B2 exceeded
B2 exceeded B1 at
B1 at of 0.84, and culminated with of 0.87. B3, B4 and B5 culminated at values
stdNDVI of 0.84, and culminated with stdNDVI of 0.87. B3, B4 and B5 culminated at stdNDVI values
ranging between 0.91 and 0.96. Above 0.96, corresponded with the highest probability of
ranging between 0.91 and 0.96. Above 0.96, stdNDVI corresponded with the highest probability of
observing B6, which represented the end of the bud burst process (Figure 5).
observing B6, which represented the end of the bud burst process (Figure 5).
S1–S3 showed the highest probability of being observed at values between 0.97 and 1.
S1–S3 showed the highest probability of being observed at stdNDVI values between 0.97 and 1.
In autumn, S4 culminated at of 0.95. At < 0.93, S5 exceeded S4, indicating that the buds
In autumn, S4 culminated at stdNDVI of 0.95. At stdNDVI < 0.93, S5 exceeded S4, indicating that the
buds were more likely to be dormant. Consequently, the bud set process was considered complete
when stdNDVI was below 0.9 (Figure 5).
Remote Sens. 2019, 11, x FOR PEER REVIEW 6 of 17

Remote Sens. 2019, 11, x FOR PEER REVIEW 6 of 17

were more likely to be dormant. Consequently, the bud set process was considered complete when
Remote Sens. 2019, 11, 2745 6 of 16
was below 0.9 (Figure 5).
were more likely to be dormant. Consequently, the bud set process was considered complete when
was below 0.9 (Figure 5).

Figure 3.
Figure 3. Correspondence between phenological
Correspondence between phenological phases
phases measured
measured in in the
the field
field and
and mean
mean standardized
standardized
MODIS
MODIS NDVI
NDVI values
values of
of black
black spruce
spruce stands
stands used
used for
for model
model calibration
calibration (n
(n == 40).
40). Dots
Dots and error
and error bars
bars
Figure 3. Correspondence between phenological phases measured in the field and mean standardized
represent mean
representNDVI and
meanvaluesstandard
and standard deviation, respectively, for the three years used for calibration.
MODIS of blackdeviation, respectively,
spruce stands used forfor the three
model years used
calibration for calibration.
(n = 40). Dots and error bars
represent mean and standard deviation, respectively, for the three years used for calibration.

Figure 4. Relationship between observed DOY and simulated DOY in three years. The black line is
Figure 4. Relationship between observed DOY and simulated DOY in three years. The black line is
the regression line between the observed DOY and the simulated DOY. The small panel shows the
the regression
Figure line between
4. Relationship the observed
between observed DOY
DOY and
and simulated
the simulated
DOYDOY. The years.
in three small The
panelblack
shows
linethe
is
distribution of
distribution of the
the residuals.
residuals.
the regression line between the observed DOY and the simulated DOY. The small panel shows the
distribution of the residuals.

Figure 5. Probability of occurrence of phenological phases derived from field observations and
standardized MODIS NDVI values of boreal black spruce stands in Quebec, Canada.
Figure 5. Probability
Probability of
of occurrence
occurrence of
of phenological
phenological phases
phases derived
derived from
from field observations and
standardized MODIS NDVI values of boreal
boreal black
black spruce
spruce stands
stands in
in Quebec,
Quebec, Canada.
Canada.
Remote Sens. 2019, 11, 2745 7 of 16

3.3. Spatial Pattern of Black Spruce Phenology

The dates of bud break (B1) were not homogenous across the study area (Figure 6). It generally
occurred earlier in the southwestern region (end of April to May) and later in the northeastern
region (mid to end of June), with the larger variations observed along the longitudinal gradient.
Overall, the earliest and latest bud break dates were observed in 2010 (DOY 165) and 2012 (DOY 180),
respectively. The highest spatial heterogeneity in bud break dates across the study area was observed
in 2012, as shown by the large variances in Table 2. Bud break was more homogenous in 2013. The
earliest bud break occurred in the last week of April (DOY 115–120) during 2010, 2012 and 2013.

Table 2. Variability in bud phenology across the study area.

Bud Break Bud Set

(DOY) (DOY)
Year
Coefficient Coefficient
Min Max Min Max
of Variation of Variation
2009 120 179 5.01 214 308 6.79
2010 115 165 4.96 216 293 4.06
2011 126 177 5.04 214 302 4.29
2012 118 180 5.80 213 309 5.22
2013 118 162 4.39 215 305 5.38
2014 121 172 4.57 211 303 4.73
2015 122 174 5.40 217 305 5.13
2016 124 172 3.02 217 309 5.66
2017 126 171 3.69 221 301 3.18
2018 126 173 5.15 216 290 3.45

On average, bud set (S5) started at the beginning of August (DOY 215) and ended at the end
of October (DOY 303) (Figure 7). The spatial pattern of bud set was not constant and no clear trend
could be found among years. Late bud set occurred more frequently in western regions, mainly in
2009 and 2014. When comparing the years, the bud set date was more homogeneous across the study
area in 2017 and more heterogeneous in 2009 (Table 2). The earliest bud set were estimated in 2013,
2016, and 2018. The range between the first and last bud break was 51 days, which is almost half
the range observed in bud set (87 days). Therefore, the spatial pattern of the growing season length
seems to be dominated by bud set. The longest growing seasons were observed in 2012 and 2014,
while the shortest ones were observed in 2013 and 2016. The southwestern and central-eastern regions
showed the longest growing seasons, while the shortest were detected in the north-central region of
the study area.
Remote Sens. 2019, 11, x FOR PEER REVIEW 9 of 17
Remote Sens. 2019, 11, 2745 8 of 16

Figure 6. Hotspot
Figure analysis
6. Hotspot of bud
analysis break
of bud (phase
break B1)B1)
(phase dates (DOY)
dates across
(DOY) boreal
across black
boreal spruce
black stands
spruce in in
stands
thethe
study area
study in in
area Quebec, Canada.
Quebec, Canada.
Remote Sens. 2019, 11, x FOR PEER REVIEW 10 of 17
Remote Sens. 2019, 11, 2745 9 of 16

Figure 7. Hotspot
Figure analysis
7. Hotspot of bud
analysis setset
of bud (phase S5)S5)
(phase dates (DOY)
dates across
(DOY) boreal
across black
boreal spruce
black stands
spruce in the
stands in the
study area in Quebec, Canada.
study area in Quebec, Canada.
Remote Sens. 2019, 11, 2745 10 of 16

4. Discussion
Time series of satellite-derived vegetation indices are a reliable tool to describe the patterns of bud
break and bud set dates across large and remote regions. In this study, we used field observations of all
black spruce phenology phases to calibrate time series of MODIS NDVI, which we used to spatialize
bud break and bud set dates across the black spruce stands in Quebec, Canada. The first phase of
bud break (B1), occurring around mid-May, corresponded to a standardized NDVI of 80.5%. NDVI
culminated at the end of July when the process of bud set was occurring. Winter bud (i.e., the last
phase of bud set) was reached when standardized NDVI reached 92.2% of its maximum amplitude
at the beginning of September. Our study provided a new statistical approach based on ordinal
logit modelling to statistically connect the sequential events of phenological phases with fortnightly
NDVI data.
Remotely sensed phenology of the boreal forest has been performed across North America or in
combination with the entire northern hemisphere [9,19,22,39]. Wide area coverage provided by satellite
remote sensing offers a unique point of view at synoptic scale but lacks in fine spatial details. To our
knowledge, species–specific calibrations of the major parameters of boreal forest bud phenology based
on satellite imagery are still missing. Fu et al. [40] calibrated MODIS vegetation phenology at global
scales (i.e., Global Vegetation Phenology) using only remote sensing phenological observations from
2001 to 2010, observing the summer culmination in NDVI and late beginning of the growing season
(in mid-April to mid-May) in the boreal and cool regions of North American evergreen coniferous
forests. White et al. [27] calibrated NDVI at 8-km resolution using ground phenology records from
1982 to 2006 for North American forests but they explicitly removed evergreen species from their
analysis. Our results improve the phenological description by calibrating MODIS NDVI time series on
field observations of bud phenology at higher spatial resolution (250 m), enhancing the reliability of
the estimations at the stand level for two important phenological phases, the beginning and end of
bud activity.
Other studies have used satellite-based retrievals of solar-induced fluorescence (SIF) to study
phenological changes in evergreen canopies [41,42]. However, apart from the spatial resolution of the
current satellites used for SIF (from ~3 km2 (OCO-2) to >1500 km2 (GOME-2)), the main difference
between the approach presented here and SIF-based phenology resides is their respective definitions
of phenological events. For example, SIF, which is mostly driven by plant fluorescence yield and by
the amount of incoming photosynthetically active radiation (PAR), is an indicator of plant carbon
uptake (gross primary production, photosynthetic light-use efficiency). Consequently, SIF- and other
phenological indicators based on plant physiology [43,44] define, for example, the start of the growing
season (SOS) as the beginning of photosynthetic activity (e.g., date of spring recovery in evergreen trees).
In contrast, with our approach, SOS is defined as the bud break date which, from a spectral perspective,
will result in visible changes in canopy color, namely greenness. Seasonally, photosynthetic activity
and changes in canopy greenness and structure are not well synchronized, either in the spring [45–47]
or in the fall and as a result, this may lead to differences in estimates of the growing season length
between physiology-based and structure-based indicators of plant phenology [42].
Compared with autumn, we observed higher slopes of the double-logistic function in spring.
The estimated winter NDVI (coefficient min of the function) was able to reduce the influence of snowmelt
during winter, thus representing an effective greening up of evergreen trees during spring [20,48].
In addition, bud break in black spruce starts at the end of May when snowmelt is completed [30].
These results are in accordance with a previous study that used NDVI time series from satellite
remote sensing to investigate xylem growth and timing of plant phenology across the boreal forests of
Quebec [49]. In this study, we demonstrated the existence of a relationship between black spruce bud
phenology phases at a wide geographical scale and a remote sensing-derived vegetation index (NDVI).
Our results show that establishing a link between bud phenology observations and NDVI time series
provides a comprehensive view of boreal forest activity patterns and trends in remote areas where
direct observations or recurrent samplings are unachievable.
Remote Sens. 2019, 11, 2745 11 of 16

Methods to detect the beginning and end of the growing season from NDVI are already available
in the literature [13,50]. In optical remote sensing, the variability in spectral signatures is considered as a
noise disturbance. In general, these noises are related to varying atmospheric conditions, the presence of
snow on coniferous canopies and the spatial and spectral variability of understorey vegetation [51–53].
Phenological events may therefore be harder to estimate for boreal evergreen coniferous forests, which
also exhibit lower seasonal changes in foliage biomass and thus, in vegetation indices such as the
NDVI [4,20].
In this study, the double-logistic model was able to suitably describe the seasonal pattern of NDVI,
despite the wide spectral variability within and between forest stands. It is well known that such a
function can effectively reduce the noise in remote sensing time series, thus adequately describing the
timings and duration of the growing season [53,54]. In previous studies, the inflection points of double
logistic models fitted to NDVI data more accurately estimated spring than autumnal events in boreal
species [20,22]. This was attributed to the longer and slower decrease in canopy greenness occurring in
autumn [4,55,56]. In this study, we were able to adequately describe and detect the inflection point in
autumn using the double-logistic function that represents the ending phase of bud set. Over the last
two decades, NDVI has been the most commonly used vegetation index for remote sensing-based
modelling of phenology [13,17,27,56]. Our method has great potential to be applied to a wider range
of species worldwide, after species–specific calibration. By providing a novel and reliable statistical
approach, this study could allow the improvement of the performance of phenological models for
evergreen forests to the same levels as has been achieved for grasslands and deciduous forests [57,58].

4.1. Relationship between NDVI and Field-Based Phenology

Our study assesses the explicit link between satellite imagery and phenological events recorded
in the field [13,59]. We observed a quick increase in NDVI during the bud break process from mid-May
to the end of June and a slow decrease in NDVI during the bud set phases from July to October.
The punctual phases of bud break were identified along the continuous process of gradual variation in
greenness. The variability in NDVI at bud set was lower than that at bud break because NDVI was
close to its maximum when bud set was already occurring. By comparing bud phenological phases
with NDVI curves, the peak standardized NDVI occurred at S2. It was observed four weeks and one
after the completion of bud burst and the start of bud set, respectively. This increase in NDVI could
be related to changes in needle biomass resulting from a gradual increase during bud burst phases
and shoot elongation during spring and summer. Furthermore, leaf expansion, i.e., increment in leaf
length and width during spring, is relatively rapid due to the transition from dormancy to active
photosynthesis [60–62].
The decrease in NDVI during late summer and autumn may be associated with needle aging
and declining pigmentation [63]. Moreover, in evergreen species, some of the old needles, commonly
those with ages varying from five to seven years, change color at the end of summer and fall in
autumn. Given that MODIS vegetation indices (NDVI, EVI) are primarily sensitive to changes in
leaf chlorophyll content and structure [18,19,64], NDVI may be more representative of phenological
changes in evergreen species during the first part of the growing season (from mid-spring to summer)
than in the autumn [4,22]. This could be attributed to the reduction in photosynthetic efficiency and
the lower carbohydrate demand during dormancy [45], which results in changes in light absorbance
and reflectance. For evergreen trees, detection of these changes is difficult as they usually require
specific and very narrow wavebands, still lacking in most current orbiting sensors [48,65]. However,
in this study, the calibration of NDVI time series with phenological observations and the resulting
threshold values alleviates some of the difficulties encountered for evergreen species when using other
approaches. This calibration is specific to black spruce stands because different thresholds of NDVI
could be estimated for other tree species of the boreal forest, mainly for deciduous species. We expect
this model to be reliable for closed stands strongly dominated by black spruce. Overall, our results
suggest that satellite NDVI may be used as a large-scale complement to field observations of bud
Remote Sens. 2019, 11, 2745 12 of 16

phenology, and therefore, support the general applicability of satellite-based vegetation indices to
estimate both the beginning and end of the growing season in boreal evergreen forests, as often defined
by bud break and bud set dates.

4.2. Spatial and Temporal Changes of Bud Phenological Phases

The beginning of bud break (phase B1) appeared later in the northeastern regions, which are
also closer to sea, resulting in a shorter duration of the overall bud break process compared with
bud set. On average, bud break lasted 51 days across the study area, which covered approximately
5◦ in latitude. This trend confirms results from other studies; however, we extended our analysis
to 5000 sites compared to a previous study that analyzed altitudinal and latitudinal gradients of
xylem phenology at only six sites black spruce sites [49]. The timings of bud break were observed
from the beginning of May to the end of June and are related to a number of factors, including the
fulfilment of the requirement in winter chilling, the lengthening of photoperiod and warming in
spring temperature [66,67]. In contrast, bud set lasted longer, 87 days, and lacked a clear and constant
spatial pattern. Bud set is an important phenological event and explains most of the variation in tree
growth [68,69], while the bud set in black spruce is a typical photoperiod-dependent process [70].
In conifers, the cessation of shoot elongation and development of terminal buds indicates vegetative
maturity. It therefore seems to result from exposure to the shortening days of late summer. During this
time, the temperature remains warm so it could be a response to some endogenous signals triggered by
the shortened photoperiod [66,71]. Overall, satellite remote sensing offers the possibility of modeling
phenology by recording spectral information at regular time intervals due to their wider coverage area
and high temporal resolution [22,72]. Our model utilized the potential of time series satellite data to
provide the spatio-temporal patterns of bud phenology by calibrating NDVI across the Quebec region
of Canada from 2009 to 2018. We expect that the presented calibration approach could be tested on a
wider basis for other sites and tree species.

5. Conclusions
With the availability of limited field-based observations due to extreme weather, remote locations
and tedious sampling, validation with the help of ground observations data is a key issue in satellite
remote sensing-based phenology analyses over large areas [73]. Uncertainties increase with the
magnitude in scale mismatch between collected field data and coarse resolution remote sensing-based
observations [13]. Calibration of bud phenological events using NDVI time series on the one hand
indicates the limitations of this approach for spatio-temporal pattern analyses, but on the other hand,
also demonstrates the ability of remote sensing to upscale bud phenology over larger forested areas.
The most confounding issues are related to the selection of vegetation index suitable for the estimation
of phenological metrics during the start and end of the growing season in evergreen forests, which
exhibit smaller seasonal variations in canopy optical properties compared to deciduous forests [22,74].
In this study, we defined a new methodology that aims to retrieve the bud phenological phases for
black spruce using MODIS-NDVI time series. We aggregated NDVI time series at stand level and
used a double-logistic curve to filter and describe NDVI-time series over time. We also used a unique
set of field-based observations to model bud phenology using ordinal logit models and proposed
statistical-based thresholds to identify the bud phenological phases for a boreal coniferous species
in Quebec, Canada. These thresholds were used to generate a spatial and temporal portrait of bud
phenology on a yearly basis over large areas. We calibrated NDVI for black spruce, but different
threshold values are expected for other tree species. Thus, our study can be reliably applied to
mostly pure black spruce stands, while applications to other species should be considered carefully.
Overall, our results suggest that satellite-derived NDVI can be used as a proxy for monitoring
phenological events when suitable statistical methods are applied to calibrate remote sensing data
with field observations.
Remote Sens. 2019, 11, 2745 13 of 16

Author Contributions: S.R., G.D., J.-D.S. and M.C.P. carried out experimental design. S.K., S.R., J.-D.S. and G.D.
did data processing, analysis and interpretation. S.K. wrote the first draft. S.K., J.-D.S., G.D. and S.R. commented
and edited the manuscript. G.D., J.-D.S. and S.R. secured funding for the study.
Funding: This work was funded by Forêt d’Enseignement et de Recherche Simoncouche, Ministère des Forêts,
de la Faune et des Parcs du Québec, Centre de Recherche sur la Boréalie, and the Fondation de l’Université du
Québec à Chicoutimi.
Acknowledgments: We thank NASA Earth science data, tools, and services for MODIS NDVI data. The authors
thanks I. Allie, L. Balducci, I. Froment, P. Nadeau, V. Néron, and S. Zhang for technical support and A. Garside for
editing the English text.
Conflicts of Interest: The authors declare no conflict of interest. The funders had no role in the design of the
study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to
publish the results.

References
1. Liu, L.; Liu, L.; Liang, L.; Donnelly, A.; Park, I.; Schwartz, M.D. Effects of elevation on spring phenological
sensitivity to temperature in Tibetan Plateau grasslands. Chinese Sci. Bull. 2014, 59, 4856–4863. [CrossRef]
2. Richardson, A.D.; Bailey, A.S.; Denny, E.G.; Martin, C.W.; O’Keefe, J. Phenology of a northern hardwood
forest canopy. Glob. Chang. Biol. 2006, 12, 1174–1188. [CrossRef]
3. Hughes, L. Biological consequences of global warming: is the signal already. Trends Ecol. Evol. 2000, 15,
56–61. [CrossRef]
4. Hmimina, G.; Dufrêne, E.; Pontailler, J.Y.; Delpierre, N.; Aubinet, M.; Caquet, B.; de Grandcourt, A.; Burban, B.;
Flechard, C.; Granier, A.; et al. Evaluation of the potential of MODIS satellite data to predict vegetation
phenology in different biomes: An investigation using ground-based NDVI measurements. Remote Sens.
Environ. 2013, 132, 145–158. [CrossRef]
5. Kang, S.; Running, S.W.; Lim, J.H.; Zhao, M.; Park, C.R.; Loehman, R. A regional phenology model for
detecting onset of greenness in temperate mixed forests, Korea: An application of MODIS leaf area index.
Remote Sens. Environ. 2003, 86, 232–242. [CrossRef]
6. White, M.A.; Thornton, P.E.; Running, S.W. A continental phenology model for monitoring vegetation
responses to interannual climatic variability. Global Biogeochem. Cycles 1997, 11, 217–234. [CrossRef]
7. Richardson, A.D.; Jenkins, J.P.; Braswell, B.H.; Hollinger, D.Y.; Ollinger, S.V.; Smith, M.L. Use of digital
webcam images to track spring green-up in a deciduous broadleaf forest. Oecologia 2007, 152, 323–334.
[CrossRef]
8. Julien, Y.; Sobrino, J.A. Global land surface phenology trends from GIMMS database. Int. J. Remote Sens.
2009, 30, 3495–3513. [CrossRef]
9. Zhang, X.; Goldberg, M.D. Monitoring fall foliage coloration dynamics using time-series satellite data. Remote
Sens. Environ. 2011, 115, 382–391. [CrossRef]
10. Liu, L.; Liang, L.; Schwartz, M.D.; Donnelly, A.; Wang, Z.; Schaaf, C.B.; Liu, L. Evaluating the potential of
MODIS satellite data to track temporal dynamics of autumn phenology in a temperate mixed forest. Remote
Sens. Environ. 2015, 160, 156–165. [CrossRef]
11. Friedl, M.A.; Sulla-Menashe, D.; Tan, B.; Schneider, A.; Ramankutty, N.; Sibley, A.; Huang, X. MODIS
Collection 5 global land cover: Algorithm refinements and characterization of new datasets. Remote Sens.
Environ. 2010, 114, 168–182. [CrossRef]
12. Garrity, S.R.; Bohrer, G.; Maurer, K.D.; Mueller, K.L.; Vogel, C.S.; Curtis, P.S. A comparison of multiple
phenology data sources for estimating seasonal transitions in deciduous forest carbon exchange. Agric. For.
Meteorol. 2011, 151, 1741–1752. [CrossRef]
13. Zhang, X.; Friedl, M.A.; Schaaf, C.B.; Strahler, A.H.; Hodges, J.C.F.; Gao, F.; Reed, B.C.; Huete, A. Monitoring
vegetation phenology using MODIS. Remote Sens. Environ. 2003, 84, 471–475. [CrossRef]
14. Rouse, W.; Haas, H.; Deering, W. 20 Monitoring Vegetation Systems in the Great Plains With Erts.
In Proceedings of the Third ERTS Symposium on Significant Results Obtained from the first Earth, College
Station, TX, USA, 1 January 1974; Volume 1, pp. 309–317.
15. Tucker, C.J.; Sellers, P.J. Satellite remote sensing of primary production. Int. J. Remote Sens. 1986, 7, 1395–1416.
[CrossRef]
Remote Sens. 2019, 11, 2745 14 of 16

16. Jeganathan, C.; Dash, J.; Atkinson, P.M. Remotely sensed trends in the phenology of northern high latitude
terrestrial vegetation, controlling for land cover change and vegetation type. Remote Sens. Environ. 2014, 143,
154–170. [CrossRef]
17. Lillesand, T.; Kiefer, R.W.; Chipman, J. Remote Sensing and Image Interpretation, 7th ed.; John Wiley & Sons:
New York, NY, USA, 2015; ISBN 978-1-118-34328-9.
18. Piao, S.; Tan, J.; Chen, A.; Fu, Y.H.; Ciais, P.; Liu, Q.; Janssens, I.A.; Vicca, S.; Zeng, Z.; Jeong, S.J.; et al. Leaf
onset in the northern hemisphere triggered by daytime temperature. Nat. Commun. 2015, 6, 6911. [CrossRef]
19. Peng, S.; Piao, S.; Ciais, P.; Myneni, R.B.; Chen, A.; Chevallier, F.; Dolman, A.J.; Janssens, I.A.; Peñuelas, J.;
Zhang, G.; et al. Asymmetric effects of daytime and night-time warming on Northern Hemisphere vegetation.
Nature 2013, 501, 88–92. [CrossRef]
20. Beck, P.S.A.; Atzberger, C.; Høgda, K.A.; Johansen, B.; Skidmore, A.K. Improved monitoring of vegetation
dynamics at very high latitudes: A new method using MODIS NDVI. Remote Sens. Environ. 2006, 100,
321–334. [CrossRef]
21. Melaas, E.K.; Friedl, M.A.; Zhu, Z. Detecting interannual variation in deciduous broadleaf forest phenology
using Landsat TM/ETM+ data. Remote Sens. Environ. 2013, 132, 176–185. [CrossRef]
22. Wu, C.; Gonsamo, A.; Gough, C.M.; Chen, J.M.; Xu, S. Modeling growing season phenology in North
American forests using seasonal mean vegetation indices from MODIS. Remote Sens. Environ. 2014, 147,
79–88. [CrossRef]
23. Xu, H.; Twine, T.E.; Yang, X. Evaluating remotely sensed phenological metrics in a dynamic ecosystem model.
Remote Sens. 2014, 6, 4660–4686. [CrossRef]
24. Elmendorf, S.C.; Jones, K.D.; Cook, B.I.; Diez, J.M.; Enquist, C.A.F.; Hufft, R.A.; Jones, M.O.; Mazer, S.J.;
Miller-Rushing, A.J.; Moore, D.J.P.; et al. The plant phenology monitoring design for the National Ecological
Observatory Network. Ecosphere 2016, 7, 1–16. [CrossRef]
25. Guyon, D.; Guillot, M.; Vitasse, Y.; Cardot, H.; Hagolle, O.; Delzon, S.; Wigneron, J.P. Monitoring elevation
variations in leaf phenology of deciduous broadleaf forests from SPOT/VEGETATION time-series. Remote Sens.
Environ. 2011, 115, 615–627. [CrossRef]
26. Richardson, A.D.; Keenan, T.F.; Migliavacca, M.; Ryu, Y.; Sonnentag, O.; Toomey, M. Climate change,
phenology, and phenological control of vegetation feedbacks to the climate system. Agric. For. Meteorol.
2013, 169, 156–173. [CrossRef]
27. White, M.A.; de Beurs, K.M.; Didan, K.; Inouye, D.W.; Richardson, A.D.; Jensen, O.P.; O’Keefe, J.; Zhang, G.;
Nemani, R.R.; van Leeuwen, W.J.D.; et al. Intercomparison, interpretation, and assessment of spring
phenology in North America estimated from remote sensing for 1982-2006. Glob. Chang. Biol. 2009, 15,
2335–2359. [CrossRef]
28. Kross, A.; Fernandes, R.; Seaquist, J.; Beaubien, E. The effect of the temporal resolution of NDVI data on
season onset dates and trends across Canadian broadleaf forests. Remote Sens. Environ. 2011, 115, 1564–1575.
[CrossRef]
29. Ministère des Ressources Naturelles et de la Faune Forêt Québec Direction des inventaires forestiers (MRNF).
Normes de Cartographie Écoforestière—Troisième Inventaire Écoforestier; Direction des Inventaires Forestiers:
Québec, QC, Canada, 2009; ISBN 978-2-550-54985-7.
30. Rossi, S.; Morin, H.; Deslauriers, A. Multi-scale influence of snowmelt on xylogenesis of black spruce. Arctic,
Antarct. Alp. Res. 2011, 43, 457–464. [CrossRef]
31. Dhont, C.; Sylvestre, P.; Gros-Louis, M.C.; Isabel, N. Guide-Terrain pour L’identification des Stades de Débourrement
et de Formation du Bourgeon Apical chez L’épinette Blanche; RNC et SCF, Centre de foresterie des Laurentides:
Québec, QC, Canada, 2010.
32. Didan, K. MOD13Q1 MODIS/Terra Vegetation Indices 16-Day L3 Global 250m SIN Grid V006; NASA EOSDIS
Land Processes DAAC: Sioux Falls, SD, USA, 2015.
33. Fisher, J.I.; Mustard, J.F.; Vadeboncoeur, M.A. Green leaf phenology at Landsat resolution: Scaling from the
field to the satellite. Remote Sens. Environ. 2006, 100, 265–279. [CrossRef]
34. Derr, B. SAS Global Forum 2013 Statistics and Data Analysis Ordinal Response Modeling with the LOGISTIC
Procedure SAS Global Forum 2013 Statistics and Data Analysis. SAS Glob. Forum 2013 2013, 1–20.
35. Prislan, P.; Gričar, J.; Čufar, K.; de Luis, M.; Merela, M.; Rossi, S. Growing season and radial growth predicted
for Fagus sylvatica under climate change. Clim. Chang. 2019, 153, 181–197. [CrossRef]
Remote Sens. 2019, 11, 2745 15 of 16

36. Getis, A.; Ord, J.K. The analysis of spatial association by distance statistics. Geogr. Anal. 1992, 355, 189–206.
[CrossRef]
37. Osei, F.B.; Duker, A.A. Spatial and demographic patterns of Cholera in Ashanti region - Ghana. Int. J. Health
Geogr. 2008, 7, 1–10. [CrossRef] [PubMed]
38. Childs, B.C. Interpolating Surfaces in ArcGIS Spatial Analyst; ESRI Education Services: Redlands, CA, USA,
2004; pp. 32–35.
39. Ganguly, S.; Friedl, M.A.; Tan, B.; Zhang, X.; Verma, M. Land surface phenology from MODIS: Characterization
of the Collection 5 global land cover dynamics product. Remote Sens. Environ. 2010, 114, 1805–1816. [CrossRef]
40. Fu, Y.; Zhang, H.; Dong, W.; Yuan, W. Comparison of Phenology Models for Predicting the Onset of Growing
Season over the Northern Hemisphere. PLoS ONE 2014, 9, e109544. [CrossRef]
41. Joiner, J.; Yoshida, Y.; Vasilkov, A.P.; Schaefer, K.; Jung, M.; Guanter, L.; Zhang, Y.; Garrity, S.; Middleton, E.M.;
Huemmrich, K.F.; et al. The seasonal cycle of satellite chlorophyll fluorescence observations and its
relationship to vegetation phenology and ecosystem atmosphere carbon exchange. Remote Sens. Environ.
2014, 152, 375–391. [CrossRef]
42. Sun, Y.; Frankenberg, C.; Wood, J.D.; Schimel, D.S.; Jung, M.; Guanter, L.; Drewry, D.T.; Verma, M.;
Porcar-Castell, A.; Griffis, T.J.; et al. OCO-2 advances photosynthesis observation from space via solar-induced
chlorophyll fluorescence. Science 2017, 358, eaam5747. [CrossRef]
43. Gamon, J.A.; Huemmrich, K.F.; Wong, C.Y.S.; Ensminger, I.; Garrity, S.; Hollinger, D.Y.; Noormets, A.;
Peñuelask, J. A remotely sensed pigment index reveals photosynthetic phenology in evergreen conifers.
Proc. Natl. Acad. Sci. USA 2016, 113, 13087–13092. [CrossRef]
44. Ulsig, L.; Nichol, C.J.; Huemmrich, K.F.; Landis, D.R.; Middleton, E.M.; Lyapustin, A.I.; Mammarella, I.;
Levula, J.; Porcar-Castell, A. Detecting inter-annual variations in the phenology of evergreen conifers using
long-term MODIS vegetation index time series. Remote Sens. 2017, 9, 49. [CrossRef]
45. Ensminger, I.; Sveshnikov, D.; Campbell, D.A.; Funk, C.; Jansson, S.; Lloyd, J.; Shibistova, O.; Öquist, G.
Intermittent low temperatures constrain spring recovery of photosynthesis in boreal Scots pine forests.
Glob. Chang. Biol. 2004, 10, 995–1008. [CrossRef]
46. Hansen, J.; Beck, E. The fate and path of assimilation products in the stem of 8-year-old Scots pine
(Pinus sylvestris L.) trees. Trees 1990, 4, 16–21. [CrossRef]
47. Monson, R.K.; Sparks, J.P.; Rosenstiel, T.N.; Scott-Denton, L.E.; Huxman, T.E.; Harley, P.C.; Turnipseed, A.A.;
Burns, S.P.; Backlund, B.; Hu, J. Climatic influences on net ecosystem CO2 exchange during the transition
from wintertime carbon source to springtime carbon sink in a high-elevation, subalpine forest. Oecologia
2005, 146, 130–147. [CrossRef] [PubMed]
48. Jönsson, A.M.; Eklundh, L.; Hellström, M.; Bärring, L.; Jönsson, P. Annual changes in MODIS vegetation
indices of Swedish coniferous forests in relation to snow dynamics and tree phenology. Remote Sens. Environ.
2010, 114, 2719–2730. [CrossRef]
49. Antonucci, S.; Rossi, S.; Deslauriers, A.; Morin, H.; Lombardi, F.; Marchetti, M.; Tognetti, R. Large-scale
estimation of xylem phenology in black spruce through remote sensing. Agric. For. Meteorol. 2017, 233,
92–100. [CrossRef]
50. Wu, C.; Hou, X.; Peng, D.; Gonsamo, A.; Xu, S. Land surface phenology of China’s temperate ecosystems
over 1999-2013: Spatial-temporal patterns, interaction effects, covariation with climate and implications for
productivity. Agric. For. Meteorol. 2016, 216, 177–187. [CrossRef]
51. Wielgolaski, F.-E.; Inouye, D.W. Phenology: An Integrative Environmental Science; Schwartz, M.D., Ed.; Springer:
Dordrecht, The Netherlands, 2003; ISBN 978-94-007-0632-3.
52. Hu, B.; Inannen, K.; Miller, J.R. Retrieval of leaf area index and canopy closure from CASI data over the
BOREAS flux tower sites. Remote Sens. Environ. 2000, 74, 255–274. [CrossRef]
53. Hird, J.N.; McDermid, G.J. Noise reduction of NDVI time series: An empirical comparison of selected
techniques. Remote Sens. Environ. 2009, 113, 248–258. [CrossRef]
54. Gonsamo, A.; Chen, J.M.; Wu, C.; Dragoni, D. Predicting deciduous forest carbon uptake phenology by
upscaling FLUXNET measurements using remote sensing data. Agric. For. Meteorol. 2012, 165, 127–135.
[CrossRef]
55. Wu, C.; Peng, D.; Soudani, K.; Siebicke, L.; Gough, C.M.; Arain, M.A.; Bohrer, G.; Lafleur, P.M.; Peichl, M.;
Gonsamo, A.; et al. Land surface phenology derived from normalized difference vegetation index (NDVI) at
global FLUXNET sites. Agric. For. Meteorol. 2017, 233, 171–182. [CrossRef]
Remote Sens. 2019, 11, 2745 16 of 16

56. Liu, Y.; Wu, C.; Peng, D.; Xu, S.; Gonsamo, A.; Jassal, R.S.; Altaf Arain, M.; Lu, L.; Fang, B.; Chen, J.M.
Improved modeling of land surface phenology using MODIS land surface reflectance and temperature at
evergreen needleleaf forests of central North America. Remote Sens. Environ. 2016, 176, 152–162. [CrossRef]
57. Hufkens, K.; Keenan, T.F.; Flanagan, L.B.; Scott, R.L.; Bernacchi, C.J.; Joo, E.; Brunsell, N.A.; Verfaillie, J.;
Richardson, A.D. Productivity of North American grasslands is increased under future climate scenarios
despite rising aridity. Nat. Clim. Chang. 2016, 6, 710–714. [CrossRef]
58. Melaas, E.K.; Friedl, M.A.; Richardson, A.D. Multiscale modeling of spring phenology across Deciduous
Forests in the Eastern United States. Glob. Chang. Biol. 2016, 22, 792–805. [CrossRef] [PubMed]
59. Lu, S.; Lu, X.; Zhao, W.; Liu, Y.; Wang, Z.; Omasa, K. Comparing vegetation indices for remote chlorophyll
measurement of white poplar and Chinese elm leaves with different adaxial and abaxial surfaces. J. Exp. Bot.
2015, 66, 5625–5637. [CrossRef] [PubMed]
60. Soudani, K.; le Maire, G.; Dufrêne, E.; François, C.; Delpierre, N.; Ulrich, E.; Cecchini, S. Evaluation of the
onset of green-up in temperate deciduous broadleaf forests derived from Moderate Resolution Imaging
Spectroradiometer (MODIS) data. Remote Sens. Environ. 2008, 112, 2643–2655. [CrossRef]
61. Yang, H.; Yang, X.; Heskel, M.; Sun, S.; Tang, J. Seasonal variations of leaf and canopy properties tracked by
ground-based NDVI imagery in a temperate forest. Sci. Rep. 2017, 7, 1–10. [CrossRef]
62. Melaas, E.K.; Sulla-Menashe, D.; Gray, J.M.; Black, T.A.; Morin, T.H.; Richardson, A.D.; Friedl, M.A. Multisite
analysis of land surface phenology in North American temperate and boreal deciduous forests from Landsat.
Remote Sens. Environ. 2016, 186, 452–464. [CrossRef]
63. Yan, D.; Zhang, X.; Nagai, S.; Yu, Y.; Akitsu, T.; Nasahara, K.N.; Ide, R.; Maeda, T. Evaluating land surface
phenology from the Advanced Himawari Imager using observations from MODIS and the Phenological
Eyes Network. Int. J. Appl. Earth Obs. Geoinf. 2019, 79, 71–83. [CrossRef]
64. Balzarolo, M.; Vicca, S.; Nguy-Robertson, A.L.; Bonal, D.; Elbers, J.A.; Fu, Y.H.; Grünwald, T.; Horemans, J.A.;
Papale, D.; Peñuelas, J.; et al. Matching the phenology of Net Ecosystem Exchange and vegetation indices
estimated with MODIS and FLUXNET in-situ observations. Remote Sens. Environ. 2016, 174, 290–300.
[CrossRef]
65. Grace, J.; Nichol, C.; Disney, M.; Lewis, P.; Quaife, T.; Bowyer, P. Can we measure terrestrial photosynthesis
from space directly, using spectral reflectance and fluorescence? Glob. Chang. Biol. 2007, 13, 1484–1497.
[CrossRef]
66. Körner, C.; Basler, D. Phenology under global warming. Science 2010, 327, 1461–1462. [CrossRef]
67. Caffarra, A.; Eccel, E. Increasing the robustness of phenological models for Vitis vinifera cv. Chardonnay.
Int. J. Biometeorol. 2010, 54, 255–267. [CrossRef]
68. Lupi, C.; Morin, H.; Deslauriers, A.; Rossi, S. Xylem phenology and wood production: Resolving the
chicken-or-egg dilemma. Plant, Cell Environ. 2010, 33, 1721–1730. [CrossRef] [PubMed]
69. Rossi, S.; Morin, H.; Deslauriers, A. Causes and correlations in cambium phenology: Towards an integrated
framework of xylogenesis. J. Exp. Bot. 2012, 63, 2117–2126. [CrossRef] [PubMed]
70. D’Aoust, A.L.; Hubac, C. Phytochrome action and frost hardening in black spruce seedlings. Physiol. Plant.
1986, 67, 141–144. [CrossRef]
71. Beaulieu, J.; Perron, M.; Bousquet, J. Multivariate patterns of adaptive genetic variation and seed source
transfer in Picea mariana. Can. J. For. Res. 2004, 34, 531–545. [CrossRef]
72. Lu, L.; Kuenzer, C.; Wang, C.; Guo, H.; Li, Q. Evaluation of three MODIS-derived vegetation index time
series for dryland vegetation dynamics monitoring. Remote Sens. 2015, 7, 7597–7614. [CrossRef]
73. Schwartz, M.D.; Reed, B.C. Surface phenology and satellite sensor-derived onset of greenness: An initial
comparison. Int. J. Remote Sens. 1999, 20, 3451–3457. [CrossRef]
74. Hufkens, K.; Friedl, M.; Sonnentag, O.; Braswell, B.H.; Milliman, T.; Richardson, A.D. Linking near-surface
and satellite remote sensing measurements of deciduous broadleaf forest phenology. Remote Sens. Environ.
2012, 117, 307–321. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).